Failsafe image sensor with real time integrity checking of pixel analog paths and digital data paths

Abstract

A method of testing analog and digital paths of a pixel in a row of an imager, includes the following steps: (a) injecting first and second charges into the analog path of the pixel, wherein the first charge is in response to a light exposure, and the second charge is in response to a built-in test; (b) sampling the first and second charges to form an image signal level and a test signal level, respectively; and (c) converting, by an analog-to-digital converter (ADC), the image signal level and the test signal level to form image data and test data, respectively. The method then validates the image data based on the test data.

What is claimed: 
     
       1. A method for testing an image sensor comprising the steps of:
 first sampling a first voltage from a pixel in a row, dependent on exposure to light; 
 performing a first analog-to-digital (A/D) conversion of the first sampled voltage to provide first digital data; 
 injecting a test signal into the pixel in the row; 
 second sampling a second voltage from the pixel in the row, after injection of the test signal; 
 performing a second A/D conversion of the second sampled voltage to provide second digital data; and 
 transferring the first and second digital data to a processor for assessing validity of the first digital data, based on the second digital data. 
 
     
     
       2. The method of  claim 1  wherein
 first sampling the first voltage includes sampling a first reset reference level of the pixel and, subsequently, sampling a first image signal level of the pixel in response to the exposure to the light; and 
 second sampling the second voltage includes sampling a second image signal level of the pixel in response to the injecting of the test signal and, subsequently, sampling a second reset reference level of the pixel in the row. 
 
     
     
       3. The method of  claim 2  wherein
 performing the first A/D conversion of the first sampled voltage includes converting the first reset reference level of the pixel and, subsequently, converting the first image signal level of the pixel; and 
 performing the second A/D conversion of the second sampled voltage includes converting the second image signal level of the pixel and, subsequently, converting the second reset reference level of the pixel. 
 
     
     
       4. The method of  claim 1  including the step of:
 resetting the pixel in the row during a shutter interval, after performing the first A/D conversion; and 
 performing the second A/D conversion includes performing the second A/D conversion during the shutter interval. 
 
     
     
       5. The method of  claim 1  wherein
 injecting the test signal includes injecting a voltage level between VAAPIX and the drain node of a reset transistor disposed in a four-transistor circuit of the pixel. 
 
     
     
       6. The method of  claim 5  wherein
 the voltage level injected is a full-scale voltage level of the image sensor. 
 
     
     
       7. The method of  claim 5  wherein
 the voltage level injected is a selected voltage level per row provided by a digital-to-analog converter (DAC), and 
 the selected voltage level varies between zero and a full-scale voltage level of the image sensor, and 
 the selected voltage level is referred to VAAPIX. 
 
     
     
       8. The method of  claim 1  including the step of:
 providing a pedestal voltage to the pixel in the row for reducing clipping of an image, wherein the pedestal voltage is based on a full-scale voltage range of the image sensor; and 
 injecting the test signal includes injecting a voltage level between VAAPIX and the drain node of a reset transistor disposed in a four-transistor circuit of the pixel, wherein the injected voltage level includes a reversed polarity of the pedestal voltage. 
 
     
     
       9. The method of  claim 1  including the steps of:
 after validating the first digital data, transferring the first digital data to an off-chip processor; and 
 after invalidating the first digital data, preventing transfer of the first digital data to the off-chip processor. 
 
     
     
       10. A method of testing analog and digital paths of a pixel in a row of an imager, comprising the steps of:
 injecting first and second charges into the analog path of the pixel, wherein the first charge is in response to a light exposure and the second charge is in response to a built-in test; 
 sampling the first and second charges to form an image signal level and a test signal level, respectively; 
 converting, by an analog-to-digital converter (ADC), the image signal level and the test signal level to form image data and test data, respectively; and 
 validating the image data based on the test data. 
 
     
     
       11. The method of  claim 10  including the steps of:
 first resetting the pixel, prior to injecting the first charge in response to the light exposure, and 
 second resetting the pixel, after injecting the second charge in response to the built-in test. 
 
     
     
       12. The method of  claim 11  wherein
 first resetting the pixel, injecting the first charge, injecting the second charge and second resetting the pixel occur, respectively, at times t 0 , t 1 , t 2 , and t 3 , and 
 each of the times are in numerical time sequence. 
 
     
     
       13. The method of  claim 11  wherein
 the second resetting of the pixel occurs during a shutter period for readying the pixel for the next frame of operation. 
 
     
     
       14. The method of  claim 10  wherein
 injecting the second charge includes providing a voltage between VAAPIX and a drain node of a reset transistor disposed in a four-transistor circuit of the pixel. 
 
     
     
       15. The method of  claim 14  wherein
 the voltage provided between VAAPIX and the drain node of the reset transistor is supplied by a digital-to-analog converter (DAC), and 
 the voltage varies between zero and a full-scale voltage level of the image sensor. 
 
     
     
       16. An image sensor comprising:
 a pixel circuit including a reset transistor having a drain node connected to a switchable voltage reference level, 
 a correlated double sampling (CDS) circuit coupled to the pixel circuit for providing a reset reference level (SHR) and an image signal level (SHS), in response to charges in the pixel circuit, and 
 a test circuit for providing the switchable voltage reference level, 
 wherein when the test circuit provides a VAAPIX voltage level to the reset transistor, the CDS circuit provides a first SHR and a first SHS, in response to charges in the pixel circuit due to light exposure, and 
 when the test circuit provides a predetermined voltage level to the reset transistor, the CDS circuit provides a second SHS, in response to charges in the pixel circuit due to the predetermined voltage level. 
 
     
     
       17. The image sensor of  claim 16  wherein
 the CDS circuit is configured to provide a second SHR, in response to charges in the pixel circuit due to the predetermined voltage level, and 
 the second SHR is provided during a shutter interval for resetting the pixel circuit for the next frame of operation. 
 
     
     
       18. The image sensor of  claim 16  wherein
 the test circuit includes a digital-to-analog converter (DAC), and 
 the DAC is configured to provide the predetermined voltage level, which varies between zero and a full-scale voltage level of the image sensor, and 
 the predetermined voltage level is referred to VAAPIX. 
 
     
     
       19. The image sensor of  claim 18  wherein
 the full-scale voltage level of the image sensor is provided by a pedestal voltage to the pixel circuit for reducing clipping of an image. 
 
     
     
       20. The image sensor of  claim 16  further including:
 an on-chip processor coupled to the CDS circuit for receiving digital data, based on the first SHR, first SHS, and second SHS, and transferring validated digital data to an off-chip processor.